Economics 2013 14

Vol. 7, 2013-14 | April 11, 2013 | http://dx.doi.org/10.5018/economics-ejournal.ja.
2013-14
Banking Systems, Central Banks and International

Reserve Accumulation in East Asian Economies
Prakash Kumar Shrestha
Abstract
This paper examines changes in the balance sheets of the banking system in five East Asian
economies which were affected by the 1997 Asian Crisis. These countries have persistently
accumulated international reserves since the crisis. This paper estimates the impact of reserve
accumulation on some important balance sheet variables such as liquid assets, credit and
deposits of the banking system by applying panel data techniques. Estimates using data from
Thailand, South Korea, Malaysia, Philippines and Indonesia show that there has been robust
positive impact of reserve accumulation on the liquid assets and deposits of the banking system
after controlling for the effect of other potential variables.
JEL E58 F31 G21
Keywords International reserves; central banks; banking systems and East Asian
countries
Authors
Prakash Kumar Shrestha, Deputy Director, Nepal Rastra Bank, Kathmandu, Nepal,
praks_shrestha@yahoo.com
Citation Prakash Kumar Shrestha (2013). Banking Systems, Central Banks and International Reserve
Accumulation in East Asian Economies. Economics: The Open-Access, Open-Assessment E-Journal, Vol. 7,
2013-14. http://dx.doi.org/10.5018/economics-ejournal.ja.2013-14
Received July 15, 2012 Published as Economics Discussion Paper September 17, 2012
Revised February 28, 2013 Accepted March 21, 2013 Published April 11, 2013
Author(s) 2013. Licensed under the Creative Commons License - Attribution 3.0
conomics: The Open-Access, Open-Assessment E-Journal
1 Introduction
Analysis of the banking system has received a great interest in recent years, espe-
cially after the global financial crisis of 200708. Studies such as Kindleberger and
Aliber (2005), Reinhart and Rogoff (2009) and Gorton (2010) have demonstrated
through historical evidence that the banking system rarely escapes any crisis.1 In
most cases, the banking system has usually aggravated and amplified the crisis, and
finally suffered from it. A series of crises has proved that a balance sheet weakness
of the banking system can ignite and propagate financial crises (Allen et al., 2002).
Hence, a growing body of literature has now emphasized the importance of balance
sheets of the banking system such as Adrian and Shin (2009), Brunnermeier et al.
(2009), Mittnik and Semmler (2011). It has been now recognized that the banking
system, in fact, plays an important role in generating boom and bust cycle in the
economy by expanding and contracting credit flows. More importantly, the banking
systems balance sheets tend to be a mirror image of the economy especially when
the balance sheets of the whole economy are not readily available (Villar, 2006).
Crises are usually manifested in the buildup of substantial balance sheet prob-
lems in the banking system as seen from the current global financial crisis and
the 1997 Asian crisis. In the Asian crisis, currency and maturity mismatches in
the banking system created a fragile financial situation (Shirai, 2001).2 The accu-
mulation of short-term external debts concentrated in the banking system, taking
advantage of the financial liberalization and globalization, was one of the important
factors behind the East Asian crisis of 1997. Banks made short-term borrowing
from abroad for long-term lending domestically, mainly to the real estate sector,
which had created imbalances in the balance sheets of the banking system, thereby
contributing to the outburst of the crisis, starting from Thailand (Shirai, 2001). In
fact, the banking system was the culprit as well as a victim of the crisis in East
Asian countries (Eichengreen, 2009; Frankel, 1998).
As a lesson learned from the painful financial crisis, many East Asian economies,
including some other emerging economies, have been building up a substantial level
of international reserves in the aftermath of the crisis, outpacing traditional bench-
mark levels (IMF, 2010). Such a persistent reserve accumulation has implications
for the balance sheet of the central bank, the banking system and the economy
as a whole (Mohanty and Turner, 2006; Banchs and Mollejas, 2010). Reserve
accumulations seem to occur by running current account surplus and intervening
in the foreign exchange markets (Schularick, 2009). Foreign exchange interven-
tions by the central banks inject the liquidity, i.e. flow of funds into the banking
system unless it is sterilized. Even if it is sterilized, the size of the balance sheets of
the banking system must change with the reserve accumulation through changing
1 With growing financialization in the economy, economic crises have increasingly concentrated in
the banking system (Cook, 2008).
2 An overexposure of the US banking system in the housing sector, financed by the inflows of funds
from abroad mainly contributed to the recent global financial crisis of 2007.
www.economics-ejournal.org 2
the portfolio of the banking system.3 The central banks foreign reserves have
counterpart liabilities in the form of bonds or currency because foreign exchange
reserve build-up has to be financed either through government budget surpluses
or by printing money or through accumulation of debt (Polterovich and Popov,
2003). With this reserve build-up, foreign currency assets play an important role
in central banks balance sheets and, consequently, in monetary policy operations
(Higgins and Klitgaard, 2004; Banchs and Mollejas, 2010). One important thing to
note is that the accumulation of international reserves successfully helped emerging
countries to weather out the adverse impact of the recent global crisis of 2007.
Yet the changing balance sheet structure of the banking system as a result of the
accumulation of international reserves has not been examined. Despite some studies
on the Asian financial crisis, for example Delhaise (1998), Caprio et al. (2005),
Cook (2008), and Carney (2009), the literature has hardly covered the changing
composition of banks balance sheets in the post-crisis period compared to the
pre-crisis period in relation to the accumulation of international reserves. On the
other hand, a number of studies have developed banking models to understand the
banking behavior such as Baltensperger (1980), Stiglitz and Greenwald (2003), and
Freixas and Rochet (2008). These models are mainly of closed economy types and
have ignored the important roles of foreign exchange markets for the banking system
in an open economy context. Only a few studies such as Mohanty and Turner (2006)
and Ho and McCauley (2009) have discussed some domestic implications of reserve
accumulation by examining average bi-variate relationship and descriptive statistics
in some Asian countries. They found no strong link between reserve accumulation
and domestic private credit growth, but did not examine the impact on the liquidity
and deposits of the banking system.
In the literature of international reserves, on the other hand, it has been argued
that accumulating international reserves seems to be a rational manner to mitigate
the impact of fundamental uncertainty and different types of risks. Experience and
evidence have shown that the global financial system is prone to various types of
risks such as currency, flight, fragility, contagion and sovereignty, and interactions
among those risks (Grabel, 2003). In this context, a stock of foreign currency
reserves provides a necessary international liquidity for self-insurance i.e. the
key to self-protection (Feldstein, 1999), since local currencies cannot take over
such a role in international payments. Ocampo (2007) argues that foreign reserves
act as a collective insurance against a balance of payments crisis, when there is
a lack of effective mechanism for macroeconomic policy coordination. Feldstein
(1999), Rajan (2008), and Banchs and Mollejas (2010) also view that holding
of international reserves appears to be essential in the world of the asymmetric
monetary system for emerging and developing countries, because of the lack of a
credible international lender of last resort and monetary cooperation at the regional
level. More importantly, given that an IMF bailout cannot be guaranteed and may
3 In recent years, emerging countries have been increasing sterilization (Aizenman and Glick, 2009).
not be adequate,4 sufficiently large precautionary reserves are felt to be essential for
a successful defense of the domestic currency and financial systems. As a result,
exchange rate stability as well as financial stability could be maintained (Stiglitz
and Greenwald, 2010; Obstfeld et al., 2010; Kato et al., 2009; Hviding et al., 2004).
Moreover, Aizenman (2006) argues that the foreign currency reserves can also
play an important role as lender of last resort in foreign currencies and the
mitigation of a terms of trade shock . Some studies such as Rodrik and Velasco
(1999), Edwards (2004), and Garca and Soto (2004) found that the probability of
capital account reversal declined with the holding of sufficient foreign exchange
reserves. It is observed that countries with a large foreign currency reserve are less
likely to be come under the currency attack (Feldstein, 1999; Cheung and Qian,
2009). On the other hand, Rodrik (2006) argues that emerging countries are paying
huge social costs by holding the large volume of international reserves. There are
some other studies which analyze the impact of capital flows on asset prices such
as stock prices and real estate, for example Caballero and Krishnamurthy (2006),
and Kim and Yang (2009). However, these studies do not touch upon the impact on
the domestic banking system from international reserve accumulation. Hence, this
paper, by departing from these arguments, attempts to contribute to the literature
by finding a new channel through which accumulation of international reserves can
help maintain financial stability in the economy through having various implications
for the banking system.
This paper seeks to investigate whether the accumulation of international re-
serves has any impacts on the balance sheet of the banking system. This will help
to draw policy implications for maintaining financial stability and find out a new
role of international reserves in the economy. This paper has taken five East Asian
countries Thailand, South Korea, Malaysia, Philippines and Indonesia affected
by the Asian crisis to examine the response of the banking system in the aftermath of
the financial crisis.5 This paper has particularly investigated the relationship of the
selected balance sheet variables of the banking system such as liquid assets, private
sector credit and deposits with the international reserve accumulation by the central
bank using panel data technique, incorporating other potential explanatory variables.
It is hypothesized that international reserve accumulation must have impact on the
banking system. These selected balance sheet variables are important for financial
stability because they can create boom-bust cycles in the economy.
The rest of the paper consists of three sections. Section 2 presents a theoretical
framework that links the central bank and the banking system, and identifies poten-
tial variables that can affect the banking system. Section 3 presents the empirical
findings and Section 4 concludes the discussion.
4 Being constrained by the limited capital, the IMF cannot provide adequate international liquidity
(Feldstein, 1999).
5 These countries have highly bank-based financial systems (see Subhanij, 2010).
2 Theoretical Framework
The banking sectors literature has mainly covered the role of the banking systems
in credit markets and deposit markets (Bhattacharya and Thakor, 1993) and banks
liquidity preference (Piegay, 2000) so far, ignoring the growing role of the banking
system in the foreign exchange markets with the globalization process. Since the
balance sheets of the central bank and the banking system are inter-connected,
international reserve accumulation by the central bank should have serious impli-
cation for the banking system. The money multiplier theory of money supply can
establish such an interconnection. For this theory, by definition, money supply in
the economy is defined as
M =C+D (1)
where M is the money supply, C is the currency held by the public, and D is
the deposits in the banking system. In the current fractional banking system, the
banking system has to keep a fraction of deposits with the central bank as a required
reserve RR. In addition, the banking system also maintain excess reserves (ER) for
smoothing payment systems as a buffer stock. On the other hand, the monetary base
or reserve money (RM) is money held by the public in currency (C) and by banks
as reserves (R) with the central bank as
RM = C + R (2)
According to the money multiplier theory of money supply, money stock in the
economy is a multiple of reserve money as
M = mRM (3)
By combining equations (1), (2) and (3), we get
C + D = m(C + R) (4)
In equation (4), D is the balance sheet item of the banking system and R is the
balance sheet item of both. In this way, equation (4) links the balance sheet of
central bank and the banking system. Table 1 presents a typical balance sheet of the
central bank. The central banks assets consist of foreign currency and domestic
assets, while its liabilities comprise currency, banking systems reserves, securities,
other liabilities and equity capital. The currency and banking systems reserves are
monetary liabilities, while the other items in the liability side are non-monetary
liabilities.
Table 1: A Typical Central Bank Balance Sheet
Assets Liabilities
Net Foreign Assets (NFA) Monetary Liabilities
Domestic Assets(DA) (i) Currency in Circulation (C)
(i) Government Securities (ii) Banking systems Reserves (R)
(ii) Loans to commercial banks Non-Monetary liabilities CBC)
(iii) Other Domestic Assets (i) Central bank securities
(ii) others
(iii) Equity Capital
The balance sheet of the central bank is written as
R +C +CBC = DA + e.NFA (5)
where R denotes the banking systems reserves with the central bank, C denotes the
currencies held by the public, CBC is the non-monetary liability of the central bank.
On the asset side, DA denotes the total domestic assets of the central bank and NFA
denotes the net foreign assets6 and e is the exchange rate7 . Then, replacing C + R in
equation (4) from equation (5) gives us
C + D = m(DA + eNFA CBC) (6)
Since the NFA consists of international reserves (IR) and other foreign assets
(e.g. gold) net of foreign liability (ONFA), the equation (6) becomes
C + D = m(DA + eIR + e.ONFA CBC) (7)
Equation (7) shows several implications from change in IR. However, assuming
other things remaining the same, one possibility could be
D
>0
IR
Further, Table 2 presents a simplified aggregate balance sheet of the banking
system. Major asset side items of the balance sheet of the banking system are private
sector credit (PC), net foreign assets (NFAb ), investment (on bond and securities),
and reserve balances with the central bank (R). On the other hand, major liability
6 The flows in the NFA represents the interactions of three sets of factors: i) foreign exchange
interventions ii) aid receipts by the Government, and iii) interest income generated by foreign currency
assets itself (Jadhav et al., 2003). However, international reserves accumulation through foreign
exchange interventions constitute the major proportion of NFA.
7 domestic currency price of foreign currency
Table 2: Simplified Aggregate Balance Sheet of the Banking System
Assets Liabilities
Private Sector Credit (PC) Deposits (D)
Net foreign assets (NFA) Other liabilities (OL)
Government & Central bank bonds (G) (i) bonds
Reserves in central bank (R) (ii) equity
side items include deposits of different types (D), and other liabilities such as bond
and equity. The balance sheet of the banking system is written as
D + OL = e.NFAb + PC + R + G (8)
where D denotes the deposits, OL denotes the other liabilities such as bonds and
capital, NFAb is the net foreign assets of the banking system, e is the exchange rate,
PC denotes the credit to the private sector, R is the reserve balance with the central
bank,8 and G is the investment on government ( and central bank) securities.
Then, replacing D by using equation (8) in equation (7), and combining R and
G as liquid assets LA, we get
PC + LA = m(DA + eIR + e.ONFA CBC) C eNFAb + OL (9)
Then, assuming other things remaining the same, there could be
PC LA
= >0 (10)
IR IR
In this way, international reserve accumulation is likely to have positive impact
on deposits, private sector credit and liquidity. This mechanism tends to work
through the foreign exchange transactions (mainly called interventions) by the
central bank with the banking system. For example, purchasing (selling) foreign
exchange injects (mops up) liquidity by increasing (reducing) the central banks
claims on nonresidents on the one hand and the banking systems reserves on the
other. If the foreign exchange interventions are sterilized through open market
operations, it increases the holding of the securities issued by the government as
well as central bank. With the change in banking systems reserves with the central
bank emanating from the foreign exchange transactions, the banking system makes
a decision on lending or changing portfolio composition.
One important thing to note here is that the nature of exchange rate system tends
to determine the necessity of foreign exchange interventions. In case of a freely
8 R broadly represents the claims of banking system on the central bank, because it also includes the
cash held by the banking system, but it does not include the central banks securities.
floating exchange rate, there should not be any flows into or out of the central banks
foreign exchange reserves. On the other hand, in case of fixed exchange rate regime,
the central bank has no freedom to control its foreign exchange reserves. In between
these two extremes, which is the situation generally observed in many emerging
and developing countries, the central bank heavily participates in foreign exchange
markets through interventions for exchange rate stability (Caprio and Honohan,
1990).9 It is indeed the foreign exchange market through which international reserve
accumulation of the central bank brings changes in the balance sheet of the banking
system. The impact of which later transmits to the other markets of the banking
system. In this way, any foreign currency inflows into the economy either from
the current account or capital account would change the balance sheets of both the
central bank and the banking system. Then, both the central bank and the banking
system actively manage their balance sheets through portfolio decisions for different
motives.10
What are the factors that affect the portfolio decision of the banking system?
The objective of the banking system is normally to maximize its profit. Following
Freixas and Rochet (2008), taking into account the management costs, the profit
() of the banking system can be written as
= rl PC + rG rD D C(D, PC) (11)
where rL is the interest rate on lending, rD is the interest rate on deposits, and r is the
interest rate on government securities and C(D, PC) is the management costs. PC,
G, and D are already defined above. The simple balance sheet identity constraint
is PC + G + R = D . Since R = D, G = (1 )D PC, where is the reserve
requirement for deposits. Then, the profit function becomes
= (rl r)PC + [r((1 ) rD ]D C(D, PC)
By taking decisions on lending and deposits, the profit maximization behavior

of the banking system is characterized by
C
= (rL r) =0
PC PC
9 Foreign exchange intervention is now considered an important balance sheet policy, which transmits
through two main channels - signaling and portfolio (Borio and Disyatat, 2009). While signaling
channel works through reflecting the central banks intention in foreign exchange interventions,
the portfolio channel brings change in relative supplies of assets (and liabilities). Since assets are
imperfect substitutes, any change in the composition of portfolios alters the behavior of banks. Under
the portfolio channel, the purchase or sale of foreign currencies from the banking system results in
a change in foreign currency portfolios of the central bank with a corresponding change in banking
systems reserves with the central bank.
10 The banks liquidity preference approach suggests that banks pursue active balance sheet policies
instead of passively accommodating the demand for credit (Bibow, 2009).
C
= [r(1 ) rD ] =0
D D
A competitive banking system adjusts its lending and deposits by making
corresponding intermediation margins equal its marginal costs. As a result, as
shown in Freixas and Rochet (2008, 73-74), dPC dD
drL > 0 and drD < 0. This means an
increase in rD will entail a decrease in the banks demand for deposits D and an
increase in rL will entail an increase in the banks supply of loans. On the other
dD
hand, the signs of dr L
and dPC
drd depend on the cross effects between deposits and
lending or the economies and dis-economies of scope. However, because of a central
role in the payment systems, the banking system cannot normally deny accepting
deposits11 but can apply discretion on lending. Further, impact of lending rate (or
deposit rate) on the holding of liquid assets can be analyzed as
LA = D PC = dLA
drL =
dD
drL dPC
drL Q 0
In addition, a competitive banking system gets equilibrium when demand for

loanable funds equals the supply of loanable funds. Let I(rL ) be the investment
demand by firms,12 and S(rD ) is the savings function of households as
I(rL ) = PC(rL , rD , r)
S(rD ) = B G + D(rL , rD , r)
where B is total government bonds issued and G is the government bonds held by
the banking system, so B G is the amount of the bonds held by the households.
Since G = (1 )D PC, S(rD ) = B + D + PC. Assuming constant marginal
0 0
costs of intermediation CL = L and CD = D , rL = r + L and rD = r(1 ) D .
In equilibrium,
(1 )
I(rL ) = [S(rD ) B PC]

which can be transformed as
1
S(rD ) I(rL ) = B (12)
1
Differentiating equation (12) with respect to B, taking into account that r as the
function of B as in Freixas and Rochet (2008), we get
11 Central bank can only ban individual banks from accepting deposits on regulatory grounds.
12 Inthis simple framework, investment demand by firm is equal to their demand for loans, since they
do not issue securities (Freixas and Rochet, 2008).
I 0 (rL )

0 dr dr
(1 )S (rD ) = 1 = >0
1 dB dB
Then, the effect on D of a change in B is now obtained by differentiating
D(rD ) = 1 [S(rD ) B I(rL )] with respect to B ( see Freixas and Rochet (2008) for
detail). Thus,
D 1
= <0
B (1 )2 S00 (rD ) 1
I (r ) L
since S0 (rD ) > 0 and I(rL ) < 0. As for the effect on PC, we can differentiate
PC = (1 )D G with respect to B. Assuming that G is a certain positive portion
(0 < < 1) of B, PC D PC
B = (1 ) B . Thus B < 0. Further, for the impact on
the liquid assets of the banking system, we differentiate
LA = D PC = D (1 )D + .B = D + .B
LA D
= +
B B
D LA
Since B < 0 , > 0, and > 0, we get B Q 0.
3 Impact of International Reserve Accumulation on the Banking Sys-

tem
This section presents empirical evidence on the impact of international reserve

accumulation on the major balance sheet variables of the banking system such
as liquid assets, credit to the private sector and deposits, by incorporating other
potential explanatory variables indicated by the theoretical framework above.
3.1 Data and Methodology

Based on the availability, annual data for the selected five East Asian economies
over the period of 1980-2010 are used for empirical estimation. The panel data
technique is applied since this technique increases the degree of freedom by pooling
the data. Since the selected countries are in the same region and follow almost
similar type of business cycle being affected by the similar external shocks, the
panel data technique seems to be appropriate. The following simple basic model is
used for empirical estimation:
0
Lit = + Xit + it , i = 1...., N; t = 1, ...., T (13)
it = i + it
E(it ) N(0, 2 )
where i and t are the country and time indices respectively; is Kx1 coefficient
vector and Xit is a vector of K explanatory variables, i denotes the unobservable
individual specific effect and vit denotes the disturbance term. Lit is the ratio of
balance sheet variables to GDP liquid assets, private sector credit and deposits in
line with Bunda and Desquilbet (2008), Vodov (2011), and Agenor et al. (2004).13
Xit is a set of explanatory variables. Based on the above theoretical discussion, Xit
includes international reserves to GDP ratio (ir), lending rate (lr) and budget deficits
as a percentage of GDP (bd) to take into account the supply of government bonds.14
Moreover, we consider the rate of GDP growth (gr), which indicates the overall
health of the economy - a higher economic growth implies a better perspective for
borrowers and more profitable lending for banks (Guo and Stepanyan, 2011) and
dummy variable for the Asian crisis period (1997 and 1998) because the selected
countries were seriously affected by that crisis. Since the balance sheet items are
closely interrelated to each other, we consider common explanatory variables for
liquidity, credit and deposits. For each dependent variable, the estimating equation
with theoretical expectation sign is written as
+ ? ?
lait = f (irit , lrit ,grit , bdit )
+ + +
pcit = f (irit , lrit ,grit , bdit )
+ ? +
d pit = f (irit , lrit ,grit , bdit )
Most of the data are taken from the IMFs International Financial Statistics
online database. However, the budget deficit data are taken from various sources
such as Asian Development Banks Key Indicator 1999 and 2010, Basic Statistics
2011, and the websites of the Central Bank of Malaysia and Indonesia, Remolona
(1985) and Nimgaonkar (2009).
Table 3 reports the descriptive statistics of the variables used in the panel
estimation. The LA to GDP ratio (la) ranges from minus 10.2 percent to 47.1
percent with 12.0 percent on average; the PC to GDP ratio (pc ) averaged at 64.8
percent, ranging from minimum 14.8 percent to maximum 165.7 percent. Similarly,
the average value of D to GDP ratio (d p ) is 61.5 percent, international reserves to
GDP (ir ) is 17.1 percent, lending rate (lr) is 12.59 percent, budget deficit to GDP
ratio (bd) is -2.04 percent, and average growth rate of GDP ( gr) is 5.6 percent (see
Table 3). Among the chosen variables, pc exhibits the highest volatility, followed
by the d p as shown by the standard deviation (SD).
13 Foruniformity across the sample countries, we use ratios of balance sheet variables in terms of
GDP.
14 Only the lending rate is chosen here assuming that different interest rates are correlated each other.
Table 3: Descriptive Statistics of the Selected Variables
la pc dp ir lr bd gr
Mean 12.03 64.75 61.46 17.09 12.59 -2.04 5.57
Max 47.13 165.72 134.71 52.81 32.15 -16.67 13.23
Min -10.18 14.85 12.07 2.05 5.02 4.75 -13.13
SD 13.04 37.68 32.36 12.53 5.63 3.17 4.22
Source: Authors calculation
100
30
pc and dp on right axis

80
20
60
10
40
0
10
20
1980 1990 2000 2010
Years of Observation
(mean) la (mean) ir
(mean) lr (mean) gr
(mean) bd (mean) pc
(mean) dp
Figure 1: Overall Movement of Variables (Mean) under Considerations
Figure 1 reports the overall movement of cross-sectional average of these

variables during the sample period. After a decline during the crisis of 1997-
1998, la exhibits an upward trend in the post-crisis period, but pc is on a downward
trend. Further, while d p and ir show an upward trend, lr exhibits a downward trend
in the post-crisis period. After a higher deficit in the 1980s, these countries had
budget surpluses or balanced budgets in the first half of the 1990s, before the crisis
hit them. After the crisis, these countries have budget deficits of varying degrees
(see Figure 1). Regarding the GDP growth (gr), it fell sharply during the crisis in
1997, recovered after that, but observed a slight fall recently in the aftermath of
the global financial crisis of 2007. Detail movement of these variables over the
whole sample period is presented in Appendix 1 and 2. Moreover, Figure 2 shows
scatter plots between the banking systems variables and international reserves. As
expected theoretically, balance sheets variables exhibits a positive correlation with
ir .
200
150
60
150
40
100
100
20
50
50
0
20
0
0 10 20 30 40 50 0 10 20 30 40 50 0 10 20 30 40 50
International reserves to GDP ratio International reserves to GDP ratio International reserves to GDP ratio
Liquid Assets to GDP ratio Fitted values Private Sector Credit to GDP ratio Fitted values Deposits to GDP ratio Fitted values
Figure 2: Scatter Plot of la, pc and d p over ir in a pooled data
Table 4: Unit Root Test
With intercept and automatic lag selection based on AIC

Null: Unit Root
adjusted t-statistics
Variable LLC test IPS test
la -0.12 (0.45) 1.14 (0.87)
dla -8.56 (0.00)* -9.64 (0.00)*
ir 2.32 (0.99) 3.28(.99)
dir -8.17 (0.00)* -8.03 (0.00)*
lr -0.43 (0.33) 0.17 (0.56)
dlr -8.59 (0.00)* -9.43 (0.00)*
bd -2.49 (0.006)* -2.85 (0.002)*
gr -6.09 (0.00)* -5.39 (0.00)*
pc -1.13(0.13) -0.58(0.28)
d pc -5.11(0.00)* -4.86(0.00)*
dp -1.18(0.12) -0.10(0.46)
d(d p) -5.68(0.00)* -6.70 (0.00)*
significant at 1%
p value in the parenthesis.
Source: Authors calculations
3.2 Empirical Results

Before estimating the model, Table 4 presents the panel unit root test proposed by
Levin et al. (2002) and Im et al. (2003), denoted by LLC15 and IPS16 respectively.
We have no prior knowledge of the number of lags, p, needed to ensure that the error
term in unit root testing equation is white noise, so we choose the number of lags
for each panel by minimizing the AIC. As shown in Table 4, variables la, pc, d p, ir
and lr have unit roots, while other variables such as gr and bd do not. However, the
first difference of la, pc, d p, ir and lr are stationary. Hence, the first difference of
these variables such as dla, d pc, dd p, dir, dlr are used for econometric estimation
later.
15 LLC test assumes all panels share a common autoregressive parameter.
16 IPS allows for individual heteroscedasticity.
There could be endogeneity between the banking systems variables, and lr

and gr. For example, a higher private sector credit i.e. pc could result in a higher
economic growth, i.e. gr, while a higher liquidity preference i.e. la could reduce the
growth of the economy through the working of the financial accelerator hypothesis.
Similarly, lending and liquidity preference behavior may induce changes in interest
rates in the banking system including lending rate. Hence, one period lag of
economic growth rate and lending rate are used, because these variables are pre-
determined and must be exogenous for current period balance sheet variables.
Looking at the last years growth and interest rate, the banking system can determine
its current period portfolio but the current portfolio decisions cannot change the
last years growth rate and interest rate. Other two variables ir and bd seem to
be exogenous for the banking system; the former depends on the current account
dynamics, capital flows and the central banks policy, and the latter depends on the
fiscal policy of the government.
An outlier is observed in Malaysia in 1990 since all banking systems variables
witnessed a substantial drop as seen in Appendix 1.17 An additional dummy variable
dum_outmal(1990 = 1) for Malaysia is hence used to remove the impact of the out-
lier. Robust standard errors are computed considering contemporaneous correlation
as well as group-wise heteroscedasticity error variance. As shown in Appendix 3
and 4, both fixed effects and random effects are not statistically significant in our
case, which shows the poolability of our data.
Table 5 presents the pooled OLS estimates. For dla, all explanatory variables
have statistically significant coefficients, with adjusted R2 = 0.52. However, only the
coefficient of gr(1) is found statistically significant for d pc with comparatively
low adjusted R2 i.e. 0.24. On the other hand, for dd p, coefficient of dir i,e. 0.65
is statistically significant at 1 % level of significance. The coefficient of gr(1)
is only significant at 10% level of significance. Adjusted R2 for dd p is relatively
higher at 0.59. The low DW Stat of 1.45 in case of d pc indicates the possibility of
serial correlation. The coefficients of dir are statistically significant in case of dla
and dd p with theoretically expected sign.
In case of time series cross-sectional data, the possibility of cross-sectional
heteroscedasticity, contemporaneous correlation and serial correlation cannot be
ruled out. Hence, Table 6 presents the feasible Generalized least square (FGLS)
estimates by correcting cross-sectional heteroscedasticity and contemporaneous
correlation. AR(1) disturbance is included for correcting serial correlation in case
of d pc and dd p. Since we have T> N, FGLS is appropriate. As we compare Table
5 and Table 6, there are some changes in coefficients of explanatory variables. In
case of dla , the explanatory power of the model i.e AdjR2 increased to 0.60, but
the significance of dlr(1) disappeared. The magnitude of coefficients of dir and
bd increased, while that of gr(1) declined. In case of d pc, only gr(1) and bd
17 The Malaysian economy recovered from the recession in the first half of 1980s and witnessed a
substantial growth of GDP in 1990. To control the economy from over heating, Statutory Reserve
Requirement (SLR) was increased gradually from 3.5 percent in 1990 to 9.5 percent in 1994.
Table 5: Pooled OLS Estimates
Cross-section SUR (PCSE) standard errors & covariance (d.f. corrected)

Dependent variables
Independent Variables
dla d pc dd p
c 1.78(0.00) -1.28(0.29) -0.68(0.43)
dir 0.53(0.00) 0.28 (0.21) 0.65(0.00)
dlr(1) 0.29(0.03) 0.10(0.63) 0.12(0.34)
gr(1) -0.29(0.00) 0.57(0.00) 0.22(0.06)+
bd -0.22(0.03) 0.43(0.11) -0.33(0.11)
dum -2.77(0.00) 0.97(0.74) 3.15(0.10)+
dum_outmal -34.94(0.00) -29.5(0.00) -68.7 (0.00)
Ad jR2 0.52 0.24 0.59
DW Stat 2.15 1.45 1.92
F stat 26.6 (0.00) 8.49(0.00) 35.26 (0.00)
N 145 145 145
+ significant at 10%; significant at 5%; significant at 1%
are found significant at 10% level of significance with the declining magnitude of
coefficients but with theoretical expected sign. The significance of AR(1) coefficient
shows the persistence nature of private sector credit. In case of dd p, as before,
dir and gr(1) are statistically significant with a declining magnitude. AR(1)
coefficient i.e. 0.18 is found significant at 5% level of significance. AdjR2 declined
from 0.59 to 0.44.
However, diagnostic checks for cross sectional independence i.e. Breusch-Pagan
LM test of cross sectional independence and White test for group-wise heteroscedas-
ticity at the bottom of Table 6 show that there is still cross sectional dependency
except in case of dd p, and group-wise heteroscedasticity in all three cases. This
may be due to the structural break and regime changes after the Asian financial
crisis. Prior to the crisis, these countries had fixed exchange rate system. After
the crisis, they adopted managed floating exchange rate, except Malaysia, which
maintained the fixed exchange rate until 2005 through capital control. All of these
countries introduced substantial banking sector reforms in the aftermath of the crisis
to strengthen the banking system. Some insolvent banks and financial institutions
were closed, and others were recapitalized and restructured (Eichengreen, 2009).
Moreover, bank supervision and regulations have been made more international
standards - high capital adequacy ratio, prudent loan loss classifications, and ad-
equate loan loss provisioning (Adams, 2008). Financing via capital markets has
been increasing as a result of the governments active involvement in developing the
bond market, which has lowered the pace of bank-led financial deepening (Adams,
Table 6: Feasible GLS Estimates with AR(1)
Number of Obs = 145

Dependent variables
dla d pc dd p
c 1.18(0.01) 0.57(0.52) -0.26(0.66)
dir 0.55(0.00) 0.15 (0.19) 0.44(0.00)
dlr(1) 0.06(0.53) -0.02(0.84) 0.06(0.49)
gr(1) -0.23(0.00) 0.13(0.09)+ 0.13(0.09)+
bd -0.31(0.00) 0.35(0.08)+ -0.18(0.28)
dum -2.10(0.02) 1.25(0.58) 3.07(0.01)
dum_outmal -35.23(0.00) -33.11(0.00) -74.15(0.00)
AR(1) 0.31(0.00) 0.18(0.04)
Ad jR2 0.60 0.24 0.44
DW Stat 2.02 1.96 2.08
F stat 37.4 (0.00) 7.18(0.00) 16.93 (0.00)
Obs 145 140 140
Diagnostic Check
LM test$ 29.80 (0.00) 42.35(0.00) 11.85 (0.29)
White Test& 110.2 (0.00) 33.6(0.03) 32.2 (0.04)
$Breusch-Pagan LM test of independence
& White Test for group-wise heteroscedasticity
2008). Hence, the relationship between dependent and independent variables might
have changed in the post-crisis period.
Hence, we perform the empirical estimates by dividing the whole sample period
into two parts pre-crisis (19801996) and post-crisis (20002010) expecting the
possible changes in relationship between independent and dependent variables. Ta-
ble 7 presents the feasible GLS estimates for the pre-crisis period. Diagnostic checks
show that there are no cross-sectional dependence and group-wise heteroscedasticity
in case of dla, but both seem to be present in case of d pc, and only group-wise
heteroscedasticity in case of dd p. Hence, the model is relatively good-fit for dla
than other two dependent variables. As we look at the estimated coefficients at Table
7, dir and bd are statistically significant in case of dla as well as dd p. Coefficients
of dlr(1) and gr(1) are significant at 10 % level of significance in case of d pc.
Table 8 presents the FGLS estimates for the post-crisis period. Diagnostic
checks show that there are no cross-sectional dependence in all three cases and no
group-wise heteroscedasticity in case of dla and dd p. This shows that our model
is relatively best fit for the post-crisis period. There is still presence of group-
wise heteroscedasticity in case of d pc, which indicates that chosen explanatory
Table 7: Feasible GLS Estimates
Pre-Crisis (19801996)
Dependent variables
dla d pc dd p
c -0.77(0.11) 2.30(0.13) 0.29 (0.77)
dir 0.85(0.00) 0.17(0.43) 0.85(0.00)
dlr(1) -0.08(0.13) -0.07(0.10)+ 0.001(0.99)
gr(1) 0.01(0.85) 0.21(0.10)+ 0.19(0.12)
bd -0.45(0.00) 0.27(0.33) -0.30 (0.04)
Dum_outmal -36.4(0.00) -34.29(0.00) -73.91(0.00)
AR(1) 0.35(0.00)
Ad jR2 0.81 0.31 0.56
DW Stat 1.78 2.03 1.81
F stat 63.56(0.00) 6.15 (0.00) 17.56 (0.00)
Obs 75 70 70
Diagnostic Check
LM test$ 11.3 (0.41) 19.37 (0.04) 2.95(0.98)
White Test& 21.0 (0.14) 34.3 (0.003) 26.25(0.03)
variables are not enough for explaining the dynamics of d pc across our selected
countries. Nevertheless, Table 8 depicts interesting findings. The coefficients of
the variable dir are now significant for all three cases. It shows that a change in
international reserves to GDP ratio seems to impact all of these selected banking
systems variables. The magnitude of the coefficient is comparatively higher in
case of dla. The impact on d pc only appeared in the post-crisis period. Another
important variable impacting both dla and d pc appeared to be bd. This implies that
higher budget deficit increases dla and d pc, which may be through availability of
liquid financial instruments for the banking system and possible complementary
effect for private sector credit flows. Other explanatory variables like gr is only
found to impact dd p, not others, and lr remained insignificant for all three cases in
the post-crisis period. Such results may be due to moderate economic growth and
low interest rate sensitiveness in these economies.
Table 8: Feasible GLS Estimates
Post-Crisis (20002010)
Dependent variables
dla d pc dd p
c -0.24(0.66) -2.67(0.03) -2.52(0.00)
dir 0.51(0.00) 0.42(0.00) 0.37(0.00)
dlr(1) 0.15(0.20) 0.01(0.89) 0.07(0.50)
gr(1) 0.03(0.69) 0.08(0.44) 0.27(0.05)
bd -0.31(0.00) -0.46(0.01) -0.22(0.26)
AR(1) 0.56(0.00)
Ad jR2 0.54 0.58 0.18
DW Stat 2.18 2.31 2.12
F stat 16.60(0.00) 16.14(0.00) 4.05(0.01)
Obs 55 55 55
Diagnostic Check
LM test$ 8.15 (0.61) 11.00(0.36) 9.08(0.52)
White Test& 19.25(0.16) 25.85(0.03) 17.6(0.23)
4 Conclusion
This paper has examined the impact of the international reserve accumulation by the
central banks on the banking systems of five East Asian countries Thailand, South
Korea, Malaysia, Philippines and Indonesia which were seriously affected by the
1997 financial crisis and which have been substantially accumulating international
reserves since then. A panel data estimation shows that international reserves
to GDP ratio (ir) is a robust factor as expected theoretically, contributing to an
increase in liquid assets and deposits of the banking system, even after taking the
possible impacts of other potential variables in both pre- and post-crisis period. In
the post-crisis period, the ir is even found to contribute positively to private sector
credit. However, our models pass the diagnostic checks for the post-crisis sample
period only, particularly for liquid assets and deposits. Hence, it can be concluded
that reserve accumulation by the central bank generates domestic liquidity and
deposits in the banking system. This type of impact of international reserves can be
considered as an additional channel for maintaining financial stability through the
accumulation of international reserves.
The holding of liquid assets, indeed, provides a cushion to withstand any
external shock to the banking system. This may be one reason that has made
the banking system of these countries resilient to the recent global financial crisis
of 2007. By contrast, these countries faced serious liquidity crunches during the
Asian crisis period because of the outflow of international reserves. Hence, the
health of the banking system in emerging countries is closely linked with the
holding of international reserves by central bank. In other words, the accumulation
of international reserves tends to contribute to financial stability by generating
domestic liquidity and creating deposits in the banking system. Because of this, an
important policy implication is that any serious unwinding of global imbalances
may have serious negative impacts on the banking systems of emerging countries.
In addition, the possibility of credit expansion cannot be ruled out, something for
which central bank should also be aware of, because the banking system can expand
lending if credit demand emerges in the economy.
Moreover, our empirical results show that the budget deficits also impact the
liquidity of the banking system positively, which may be due to the fact that gov-
ernment deficit helps to develop liquid bond markets. In addition, since emerging
countries have space for fiscal policy, budget deficits can play a complementary
role in the expansion of private sector credit, as found in empirical results in the
post-crisis period. One policy implication of such a finding is that in addition to
other macroeconomic effects, government budget deficits could have significant
implications for the banking system.
This study can be extended in several directions. First, the panel estimation
can be done by extending the sample countries, which could include not only
Asian countries but also Latin American and African countries. Second, other
controlling variables could be incorporated to examine the robustness of the impact
of international reserves on the balance sheet variables of the banking system such
as capital adequacy ratio and non-performing assets. Third, the seemingly unrelated
regression (SUR) technique could be applied to verify the empirical results.
Acknowledgments
This article is a product of my PhD dissertation submitted to the New School for Social
Research, New York. I would like to acknowledge the financial support provided by The
New School and Open Society Foundation for this research. More importantly, I am grateful
to two anonymous referees for providing valuable comments and suggestions, which have
improved the paper. The views expressed here are our own and do not reflect the affiliated
institution.
References
Adams, C. (2008). Emerging East Asian Banking Systems Ten Years

after the 1997/98 Crisis. ADB Working Paper Series on Regional
Economic Integration 16, Asian Development Bank, Manila. URL
http://www.adb.org/documents/papers/regional-economic-integration/
WP16-Emerging-EastAsian-Banking-Systems.pdf.
Adrian, T., and Shin, H. (2009). Money, Liquidity, and Monetary Policy. American
Economic Review, 99(2): 600605. URL http://ideas.repec.org/a/aea/aecrev/
v99y2009i2p600-605.html.
Agenor, P.-R., Aizenman, J., and Hoffmaister, A. (2004). The Credit Crunch in East
Asia: What can Bank Excess Liquid Assets Tell us? Journal of International
Money and Finance, 23(1): 2749. URL http://www.sciencedirect.com/science/
article/pii/S026156060300069X.
Aizenman, J. (2006). International Reserves Management and the Current Ac-

count. NBER Working Paper 12734, National Bureau of Economic Research,
Cambridge, MA. URL http://ideas.repec.org/p/nbr/nberwo/12734.html.
Aizenman, J., and Glick, R. (2009). Sterilization, Monetary Policy, and Global
Financial Intergration. Review of International Economics, 17: 777801. URL
http://ideas.repec.org/a/bla/reviec/v17y2009i4p777-801.html.
Allen, M., Rosenberg, C., Keller, C., Setser, B., and Roubini, N. (2002). A Balance
Sheet Approach to Financial Crisis. IMF Working Paper WP/02/210, International
Monetary Fund, Washington, D.C. URL http://ideas.repec.org/p/imf/imfwpa/
02-210.html.
Baltensperger, E. (1980). Alternative Approaches to the Theory of the Banking

Firm. Journal of Monetary Economics, 6: 137. URL http://ideas.repec.org/a/
eee/moneco/v6y1980i1p1-37.html.
Banchs, A., and Mollejas, L. (2010). International Monetary Asymmetries and

the Central Bank. Journal of Post Keynesian Economics, 32(3): 467496. URL
http://ideas.repec.org/a/mes/postke/v32y2010i3p467-496.html.
Bhattacharya, S., and Thakor, A. (1993). Contemporary Banking Theory. Journal

of Financial Intermediation, 3(1): 250.
Bibow, J. (2009). Keynes on Monetary Policy, Finance and Uncertainty: Liquidity

Preference Theory and the Global Financial Crisis. Routledge, London. URL
http://www.routledge.com/books/details/9780415352628/.
Borio, C., and Disyatat, P. (2009). Unconventional Monetary Policies: An Appraisal.

BIS Working Paper 292, Bank for International Settlements, Basel. URL http:
//www.bis.org/publ/work292.pdf.
Brunnermeier, M., Nagel, S., and Pedersen, L.-H. (2009). Carry Trades and Currency
Crashes. In NBER Macroeconomic Annual 2008, Vol. 23. University of Chicago
Press, Chicago, IL.
Bunda, I., and Desquilbet, J.-B. (2008). The Bank Liquidity Smile Across Exchange
Rate Regimes. International Economic Journal, 22(3): 361386. URL http:
//ideas.repec.org/a/taf/intecj/v22y2008i3p361-386.html.
Caballero, R., and Krishnamurthy, A. (2006). Bubbles and Capital Flow Volatility:
Causes and Risk Management. Journal of Monetary Economics, 53(1): 3553.
URL http://www.nber.org/papers/w11618.
Caprio, G., Hanson, J., and Litan, R. (Eds.) (2005). Financial Crises: Lessons from
the Past, Preparation for the Future. Brookings Institution Press, Washington,
D.C.
Caprio, G., and Honohan, P. (1990). Monetary Policy Instruments for Developing
Countries. The World Bank Working Paper 528, The World Bank, Washington,
D.C.
Carney, R. (Ed.) (2009). Lessons from the Asian Financial Crisis. Routledge, New
York & London.
Cheung, Y.-W., and Qian, X. (2009). Hoarding of International Reserves: Mrs

Machlups Wardrobe and the Joneses. Review of International Economics, 17(4):
824843. URL http://ideas.repec.org/a/bla/reviec/v17y2009i4p824-843.html.
Cook, M. (2008). Banking Reform in Southeast Asia: The Regions Decisive

Decade. Routledge, Abingdon. URL http://www.crcnetbase.com/doi/pdf/10.4324/
9780203926833.fmatt.
Delhaise, P. (1998). Asia in Crisis. John Wiley & Sons (Asia) Pte Ltd., Singapore-
New York.
Edwards, S. (2004). Thirty Years Of Current Account Imbalances, Current Account

Reversals and Sudden Stops. NBER Working Paper 10276, National Bureau of
Economic Research, Cambridge, MA. URL http://ideas.repec.org/p/nbr/nberwo/
10276.html.
Eichengreen, B. . (2009). The Asian Crisis After 10 Years. In R. Rajan,

S. Thangavelu, and R. Parinduri (Eds.), Exchange Rate, Monetary and Financial
Issues and Policies in Asia. World Scientific, Singapore.
Feldstein, M. (1999). A Self-Help Guide for Emerging Markets. Foreign

Affairs, 78(2): 93109. URL http://www.foreignaffairs.com/articles/54801/
martin-feldstein/a-self-help-guide-for-emerging-markets.
Frankel, J. (1998). The Asian Model: The Miracle, The Crisis and The Fund. The
U.S. International Trade Commission, Washington, D.C. URL http://www.hks.
harvard.edu/fs/jfrankel/eacritc.pub.pdf.
Freixas, X., and Rochet, J.-C. (2008). Microeconomics of Banking. MIT Press,
Cambridge, MA.
Garca, P., and Soto, C. (2004). Large Hoardings of International Reserves: Are
they Worth it? Central Bank of Chile Working Paper 299, Central Bank of Chile,
Santiago. URL http://www.bcentral.cl/estudios/documentos-trabajo/pdf/dtbc299.
pdf.
Gorton, G. (2010). Slapped by the Invisible Hand: The Panic of 2007. Oxford
University Press, New York.
Grabel, I. (2003). Averting Crisis? Assessing Measures to Manage Financial

Integration in Emerging Economies. Cambridge Journal of Economics, 27(3):
317336. URL http://cje.oxfordjournals.org/content/27/3/317.abstract.
Guo, K., and Stepanyan, V. (2011). Determinants of Bank Credit in Emerging

Market Economies. IMF Working Paper 11/51, International Monetary Fund,
Washington, D.C. URL http://www.imf.org/external/pubs/cat/longres.aspx?sk=
24698.0.
Higgins, M., and Klitgaard, T. (2004). Reserve Accumulation: Implications for

Global Capital Flows and Financial Markets. Current Issues in Economics
and Finance, 10(10): 18. URL http://newyorkfed.org/research/current_issues/
ci10-10.pdf.
Ho, C., and McCauley, R. (2009). The Domestic Financial Consequences of Reserve
Accumulation: Some Evidence from Asia. In R. Rajan, S. Thangavelu, and
R. Pariduri (Eds.), Exchange Rate, Monetary and Financial Issues and Policies
in Asia. World Scientific, Singapore.
Hviding, K., Nowak, M., and Ricci, L. (2004). Can Higher Reserves Help Reduce
Exchange Rate Volatility? IMF Working Paper 04/189, International Monetary
Fund, Washington, D.C. URL http://www.imf.org/external/pubs/ft/wp/2004/
wp04189.pdf.
Im, K., Pesaran, M., and Shin, Y. (2003). Testing for Unit Roots in Heterogeneous
Panels. Journal of Econometrics, 115(1): 5374. URL http://ideas.repec.org/a/
eee/econom/v115y2003i1p53-74.html.
IMF (2010). Reserve Accumulation and International Monetary Stability. Interna-

tional Monetary Fund, Washington, D.C. URL http://www.imf.org/external/np/
pp/eng/2010/041310.pdf.
Jadhav, N., Ray, P., Bose, D., and Gupta, I. S. (2003). The Reserve Bank of Indias
Balance Sheet: Analytics and Dynamics of Evolution. Reserve Bank of India
Occasional Papers, 24(3): 159. URL http://www.rbi.org.in/upload/Publications/
PDFs/60611.pdf.
Kato, M., Proao, C., and Semmler, W. (2009). Currency Runs, International
Reserves Management and Optimal Monetary Policy Rules. IMK Working Paper
8/2009, Hans Boeckler Foundation, Macroeconomic Policy Institute, Dsseldorf.
URL http://www.boeckler.de/pdf/p_imk_wp_8_2009.pdf.
Kim, S., and Yang, D. (2009). Do Capital Inflows Matter to Asset Prices? The Case
of Korea. Asian Economic Journal, 23(3): 323348. URL http://onlinelibrary.
wiley.com/doi/10.1111/j.1467-8381.2009.02014.x/abstract.
Kindleberger, C., and Aliber, R. (2005). Manias, Panics and Crashes: A History
of Financial Crises. Palgrave Macmillan, Houndsmill, Basingstoke, Hampshire,
UK.
Levin, A., Lin, C.-F., and Chu, C.-S. (2002). Unit Root Tests in Panel Data:
Asymptotic and Finite-Sample Properties. Journal of Econometrics, 108(1):
124.
Mittnik, S., and Semmler, W. (2011). The Instability of the Banking Sector and
Macrodynamics: Theory and Empirics. CEM Working Paper February, Com-
parative Empirical Macroeconomics, Department of Economics, New School
University, New York. URL http://www.newschool.edu/cem/financial-markets/.
Mohanty, M., and Turner, P. (2006). Foreign Exchange Reserve Accumulation in

Emerging Markets: What are the Domestic Implications? BIS Quarterly Review,
September: 3953. URL http://www.bis.org/publ/qtrpdf/r_qt0609f.pdf.
Nimgaonkar, G. (2009). Export Led Growth in South East Asia: An Overview of

Export Policies and Lessons For India. In 9th Global Conference in Business
and Economics, Cambridge University, UK. URL www.gcbe.us/9th_GCBE/data/
Girija%20Nimgaonkar.doc.
Obstfeld, M., Shambaugh, J., and Taylor, A. (2010). Financial Stability, the
Trilemma, and International Reserves. American Economics Journal: Macroeco-
nomics, 2(2): 5794. URL http://ideas.repec.org/a/aea/aejmac/v2y2010i2p57-94.
html.
Ocampo, J. (2007). The Instability and Inequities of the Global Reserve System.
DESA Working Paper 59, Department of Economic and Social Affairs, United
Nations, New York. URL http://www.un.org/esa/desa/papers/2007/wp59_2007.
pdf.
Piegay, P. (2000). New and Post Keynesian Analyses of Bank Behavior: Consensus
and Disagreement. Journal of Post Keynesian Economics, 22(2): 265283. URL
http://www.jstor.org/stable/4538679.
Polterovich, V., and Popov, V. (2003). Accumulation of Foreign Exchange Reserves

and Long Term Growth. MPRA Paper 20069, University Library of Munich,
Munich. URL http://mpra.ub.uni-muenchen.de/20069/.
Rajan, R. (2008). Monetary and Financial Cooperation in Asia: Talking Stock

of Recent On-goings. International Relations of the Asia-Pacific, 8: 31
45. URL http://ramkishenrajan.gmu.edu/pdfs/publications/Published_Academic_
Papers/2008/mfr.pdf.
Reinhart, C., and Rogoff, K. (2009). This Time Is Different: Eight Centuries of
Financial Folly. Princeton University Press, Princeton, NJ.
Remolona, E. (1985). Financing the Budget Deficit in the Philippines. Institute for
Development Studies Working Paper 85-02, Philippine Institute for Development
Studies, Makati City. URL http://dirp4.pids.gov.ph/ris/wp/pidswp8502.pdf.
Rodrik, D. (2006). The Social Cost of Foreign Exchange Reserves. Interna-

tional Economic Journal, 20(3): 253266. URL http://ideas.repec.org/a/taf/intecj/
v20y2006i3p253-266.html.
Rodrik, D., and Velasco, A. (1999). Short-Term Capital Flows. In Annual World
Bank Conference on Development Economics. The World Bank, Washington,
D.C.
Schularick, M. (2009). Touching the Brakes after the Crash: A Historical View of Re-
serve Accumulation and Financial Integration. Global Economy Journal, 9(4): 1
13. URL http://www.jfki.fu-berlin.de/faculty/economics/team/persons/schularick/
A_Historical_View_of_Reserve_Accumulation_and_Financial_Integration.pdf.
Shirai, S. (2001). Overview of Financial Market Structures in Asia. ADB Institute

Research Paper 25, Asian Development Bank, Manila. URL http://www.adbi.org/
files/2001.09.rp25.market.structure.asia.pdf.
Stiglitz, J., and Greenwald, B. (2003). Towards a New Paradigm in Monetary

Economics. Cambridge University Press, Cambridge.
Stiglitz, J., and Greenwald, B. (2010). Towards A New Global Reserve System.
Journal of Globalization and Development, 1(2): 126. URL http://ideas.repec.
org/a/bpj/globdv/v1y2010i2n10.html.
Subhanij, T. (2010). Liquidity Measurement and Management in the SEACEN

Economies. The SEACEN Centre; Kuala Lumpur.
Villar, A. (2006). Is Financial Stability Policy now Better Placed to Prevent Systemic
Banking Crises? BIS Paper 28, Bank for International Settlements, Basel. URL
http://www.bis.org/publ/bppdf/bispap28.pdf.
Vodov, P. (2011). Liquidity of Czech Commercial Banks and its Determinants.

International Journal of Mathematical Models and Methods in Applied Sciences,
5(6): 10601067. URL http://www.naun.org/multimedia/NAUN/m3as/20-855.
pdf.
Appendix 1: Movement of Dependent Variables
Thailand South Korea

30 15
20
10
10 120
5
0
200 100
-10 0
160 80
-20 -5
120
60
80 -10
40
40
0 20
1980 1985 1990 1995 2000 2005 2010 1980 1985 1990 1995 2000 2005 2010
Malaysia Philippines
50 30
40
30 20
20
60
160 10 10
50
0
120 40 0
30
80
20
40 10
1980 1985 1990 1995 2000 2005 2010 1980 1985 1990 1995 2000 2005 2010
Indonesia
40
30
20
10
60 0
50 -10
40
30
20
10
1980 1985 1990 1995 2000 2005 2010
LA (Right Axis) PC DP
Appendix 2: Movement of Independent Variables
(IR=International Reserves, GR = growth rate, LR=Lending Rate, BD = Budget

Deficit; IR and BD are as a percentage of GDP)
IR GR
60 15
50 10
40 5
30 0
20 -5
10 -10
0 -15
1980 1985 1990 1995 2000 2005 2010 1980 1985 1990 1995 2000 2005 2010
LR BD
35 5
30
0
25
20 -5
15 -10
10
-15
5
0 -20
1980 1985 1990 1995 2000 2005 2010 1980 1985 1990 1995 2000 2005 2010
Thailand South Korea

Malaysia Philippines
Indonesia
Appendix 3: Fixed Effects
Dependent variables
dla d pc dd p
c 1.81(0.00) -0.40(0.77) -0.23(0.82)
dir 0.53(0.00) 0.26(0.26) 0.64(0.00)
dlr(1) 0.29(0.03) 0.10(0.61) 0.12(0.33)
gr(1) -0.30(0.00) 0.50(0.02) 0.18(0.13)
bd -0.22(0.07) 0.66 (0.03) -0.22(0.35)
dum -2.76(0.00) 0.96(0.74) 3.15(0.11)
dum_outmal -35.05(0.00) -31.9(0.00) -70.03(0.00)
Ad jR2 0.50 0.25 0.58
DW Stat 2.15 1.52 1.95
F stat 15.59 5.71 21.31
N 145 145 145
Redundant Fixed Effect test
Cross section F 0.10(0.98) 1.49(0.21) 0.76(0.56)
Cross Section 2 0.42(0.98) 6.36(0.18) 3.23 (0.52)

Appendix 4: Random Effects
Dependent variables
dla d pc dd p
c 1.79(0.00) -1.11(0.38) -0.68(0.43)
dir 0.53(0.00) 0.28(0.22) 0.65(0.00)
dlr(1) 0.29(0.03) 0.10(0.63) 0.12(0.34)
gr(1) -0.29(0.00) 0.56(0.00) 0.22(0.06)
bd -0.22(0.03) 0.47(0.08) -0.33(0.12)
dum -2.77(0.01) 0.97(0.74) 3.15(0.10)
dum_outmal -34.94(0.00) -29.97(0.00) -68.73(0.00)
Ad jR2 0.52 0.24 0.59
DW Stat 2.15 1.46 1.91
F stat 26.61 (0.00) 8.55 (0.00) 35.26(0.00)
N 145 145 1.45
Hausman test 0.19(0.99) 0.00 (1.00) 0.00 (1.00)

Please note:
You are most sincerely encouraged to participate in the open assessment of this article. You
can do so by either recommending the article or by posting your comments.
Please go to:
http://dx.doi.org/10.5018/economics-ejournal.ja.2013-14
The Editor
Author(s) 2013. Licensed under the Creative Commons Attribution 3.0.

Economics 2013 14

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Economics 2013 14

Uploaded by

Copyright:

Available Formats

Vol. 7, 2013-14 | April 11, 2013 | http://dx.doi.org/10.5018/economics-ejournal.ja.

Banking Systems, Central Banks and International

By combining equations (1), (2) and (3), we get

Table 1: A Typical Central Bank Balance Sheet

The balance sheet of the central bank is written as

R +C +CBC = DA + e.NFA (5)

C + D = m(DA + eNFA CBC) (6)

C + D = m(DA + eIR + e.ONFA CBC) (7)

Table 2: Simplified Aggregate Balance Sheet of the Banking System

PC + LA = m(DA + eIR + e.ONFA CBC) C eNFAb + OL (9)

Then, assuming other things remaining the same, there could be

= rl PC + rG rD D C(D, PC) (11)

= (rl r)PC + [r((1 ) rD ]D C(D, PC)

By taking decisions on lending and deposits, the profit maximization behavior

instead of passively accommodating the demand for credit (Bibow, 2009).

In addition, a competitive banking system gets equilibrium when demand for

3 Impact of International Reserve Accumulation on the Banking Sys-

This section presents empirical evidence on the impact of international reserve

3.1 Data and Methodology

Table 3: Descriptive Statistics of the Selected Variables

pc and dp on right axis

Figure 1: Overall Movement of Variables (Mean) under Considerations

Figure 1 reports the overall movement of cross-sectional average of these

Figure 2: Scatter Plot of la, pc and d p over ir in a pooled data

Table 4: Unit Root Test

With intercept and automatic lag selection based on AIC

3.2 Empirical Results

There could be endogeneity between the banking systems variables, and lr

Table 5: Pooled OLS Estimates

Cross-section SUR (PCSE) standard errors & covariance (d.f. corrected)

Table 6: Feasible GLS Estimates with AR(1)

Number of Obs = 145

Table 7: Feasible GLS Estimates

Table 8: Feasible GLS Estimates

Adams, C. (2008). Emerging East Asian Banking Systems Ten Years

Aizenman, J. (2006). International Reserves Management and the Current Ac-

Baltensperger, E. (1980). Alternative Approaches to the Theory of the Banking

Banchs, A., and Mollejas, L. (2010). International Monetary Asymmetries and

Bhattacharya, S., and Thakor, A. (1993). Contemporary Banking Theory. Journal

Bibow, J. (2009). Keynes on Monetary Policy, Finance and Uncertainty: Liquidity

Borio, C., and Disyatat, P. (2009). Unconventional Monetary Policies: An Appraisal.

Cheung, Y.-W., and Qian, X. (2009). Hoarding of International Reserves: Mrs

Cook, M. (2008). Banking Reform in Southeast Asia: The Regions Decisive

Edwards, S. (2004). Thirty Years Of Current Account Imbalances, Current Account

Eichengreen, B. . (2009). The Asian Crisis After 10 Years. In R. Rajan,

Feldstein, M. (1999). A Self-Help Guide for Emerging Markets. Foreign

Grabel, I. (2003). Averting Crisis? Assessing Measures to Manage Financial

Guo, K., and Stepanyan, V. (2011). Determinants of Bank Credit in Emerging

Higgins, M., and Klitgaard, T. (2004). Reserve Accumulation: Implications for

IMF (2010). Reserve Accumulation and International Monetary Stability. Interna-

Mohanty, M., and Turner, P. (2006). Foreign Exchange Reserve Accumulation in

Nimgaonkar, G. (2009). Export Led Growth in South East Asia: An Overview of

Polterovich, V., and Popov, V. (2003). Accumulation of Foreign Exchange Reserves

Rajan, R. (2008). Monetary and Financial Cooperation in Asia: Talking Stock

Rodrik, D. (2006). The Social Cost of Foreign Exchange Reserves. Interna-

Shirai, S. (2001). Overview of Financial Market Structures in Asia. ADB Institute

Stiglitz, J., and Greenwald, B. (2003). Towards a New Paradigm in Monetary

Subhanij, T. (2010). Liquidity Measurement and Management in the SEACEN

Vodov, P. (2011). Liquidity of Czech Commercial Banks and its Determinants.

Appendix 1: Movement of Dependent Variables

Thailand South Korea

Appendix 2: Movement of Independent Variables

(IR=International Reserves, GR = growth rate, LR=Lending Rate, BD = Budget

Thailand South Korea